Liveness Detection
Verify the Presence of a Live Person — Not a Photo or Mask.
id3's liveness detection algorithms distinguish genuine biometric presentations from spoofing attacks in real time — printed photos, replayed videos, silicone masks and deepfakes. Passive and active modes, deployable on mobile, server and edge.
Explore Face PADWhat It Detects
Presentation Attack Instruments (PAI) covered by id3's liveness detection — across both passive and active modes.
Printed Photo
Flat print or photograph of the victim's face held in front of the camera.
Video Replay
Pre-recorded video of the victim played on a screen or phone in front of the sensor.
3D Mask
Rigid or flexible 3D mask — silicone, resin or paper — mimicking the target's facial geometry.
Deepfake / Face Swap
AI-generated face swapped onto a live video stream in real time via digital injection.
Cut-Out Attack
Printed photo with eye holes cut out, or animated paper with moving eye regions.
Digital Injection
Synthetic video stream injected directly into the camera API, bypassing the physical sensor entirely.
Detection Modes
Choose the mode that fits your UX and security requirements — or combine both for maximum assurance.
Passive Liveness
A single selfie is sufficient — no user action required. The algorithm analyzes depth cues, texture, reflection and micro-movements in a single frame or short sequence.
- No user interaction
- Works with standard cameras
- Sub-second decision
- Invisible to the user
- Photo, video & 3D mask detection
Active Liveness
The user is prompted to perform specific actions — blink, smile, turn their head — in a randomized sequence that cannot be replayed or predicted.
- Challenge-response protocol
- Randomized action sequence
- Blocks video replay attacks
- Configurable difficulty
- Multi-frame analysis
Choose Your Integration
Liveness detection is available as a standalone API or embedded in the Face SDK for native integration.
Face PAD API
REST web service for passive and active liveness detection. Integrates into any web or mobile application in minutes — no SDK, no binary dependency.
Face SDK
Native library with liveness detection built-in alongside face detection, analysis, extraction and matching — single SDK for the full pipeline.
Application Domains
Liveness detection is a critical layer in any identity verification workflow where remote or unattended authentication is required.
KYC & Digital Onboarding
Verify that the person submitting a selfie is physically present — not using a photo of a stolen ID document.
Online Banking
Secure high-value transactions and account changes with a biometric selfie that cannot be spoofed by a photo or video replay.
Access Control
Prevent photo-based attacks on face recognition access control systems — gates, doors and logical access to enterprise systems.
Mobile Authentication
Add passive liveness to any mobile app login or payment confirmation — invisible check, no user friction, strong security.
Remote Healthcare
Patient identity verification for telehealth consultations and remote prescription authorization.
e-Government Services
Biometric verification for remote civil service access — tax filing, benefit claims, driving licence renewal.
Why id3 Technologies
Presentation Attack Detection
Algorithms developed to detect printed photos, video replays and 3D masks — covering the full range of presentation attacks.
Single-Frame Passive Mode
Passive liveness works on a single image — no video stream required. Compatible with any camera including low-end mobile front cameras.
Embedded-Ready
Liveness runs inside the Face SDK on iOS, Android and edge devices — no cloud round-trip, no latency, no privacy exposure.
Algorithm-Level Expertise
id3 develops its own liveness algorithms — not a licensed third-party module. Direct access to the research team for custom tuning and threat model updates.
Frequently Asked Questions
Everything you need to know about id3's liveness detection technology.
01 What is liveness detection?
Liveness detection (also called Presentation Attack Detection or PAD) verifies that the biometric sample being presented comes from a live person — not a photo, video, mask or digital injection. It is a mandatory layer in any remote biometric authentication system.
02 What is the difference between passive and active liveness?
Passive liveness requires no user action — it analyzes depth, texture and micro-movement cues from a selfie. Active liveness prompts the user to perform a randomized action (blink, smile, head turn) to confirm they are physically present. Active mode provides stronger protection against video replay attacks.
03 Can it detect deepfakes and digital injection attacks?
Yes. id3's algorithms include signal-level analysis that detects anomalies in the video stream characteristic of digital injection and synthetic face generation, in addition to physical presentation attack detection.
04 Does it require a special camera?
No. Passive liveness works with any standard RGB camera — including low-resolution mobile front cameras. Depth cameras (ToF, structured light) can optionally be used to strengthen 3D mask detection, but are not required.
05 What types of presentation attacks does it detect?
id3's liveness detection targets printed photos, video replays, silicone and resin 3D masks, and deepfakes — using passive analysis and active challenge-response.
06 Can it integrate with existing biometric systems?
Yes. The Face PAD API integrates via REST into any existing identity verification stack. The Face SDK embeds liveness directly alongside detection and recognition — no separate service, no additional latency.
Get started with our technologies.
Contact us to learn more about our biometric and security solutions and discover how it can transform your products and services. With id3 Technologies, step into a world where technology meets security, innovation, and reliability.
Contact us